Assembly x64 Intro - SSE2 4x4D Transpose

; in: xmm1, xmm2, xmm3, xmm4, xmm5 pOut: xmm1, xmm4, xmm5, mm3

%macro SSE2_Trans4x4D 5

SSE2_XSawp dq, %1, %2, %5

SSE2_XSawp dq, %3, %4, %2

SSE2_XSawp qdq, %1, %3, %4

SSE2_XSawp qdq, %5, %2, %3

%endmacro

;for TRANSPOSE

%macro SSE2_XSawp 4

movdqa %4, %2

punpckl%1 %2, %3

punpckh%1 %4, %3

%endmacro

SSE2_Trans4x4D xmm4, xmm2, xmm1, xmm3, xmm5 ; pOut: xmm4,xmm3,xmm5,xmm1

类似 MMX_Trans4x4W， MMX_Trans4x4W操作的是16bit的字， 而SSE2_Trans4x4D 操作的是双字。